In the light of recent discussion regarding Newspaper Licensing Ireland and website links, I thought it might prove useful to look at what the law behind the whole dispute says, and at least as importantly, what it does not say.
A hyperlink direct to a particular page within a website has been termed a “deep link”, to distinguish it from a link which simply directs one to the homepage of a particular publication, (www.irishtimes.com, for e.g.). The validity of any such distinction is in fact questionable. The whole point of the internet is that it is a collection of individual pages, none any “deeper” than the other. Homepages, particularly news media organisation homepages, usually display copyright material. In a copyright context therefore, a link to a home page is no different in principle to a direct link to an article.
Links are not explicitly provided for anywhere in Irish law. When pressed in recent days, the newspaper industry could point to no specific provision to support their assertion that links require a licence. However, the most immediately relevant provisions are Sections 39 and 40 of the Copyright and Related Rights Act, 2000.
Section 40(8) provides for a “Making Available Right”:
“There shall be a right of the owner of copyright to make available to the public copies of a work or to authorise others to do so which shall be known and in this Part referred to as the ‘‘making available right’’.”
This right is defined as including making available copies of the work though the internet. The question here is whether posting a link constitutes such a “making available”. If so, it is something for which the copyright owner is entitled to withhold or charge for permission.
Robert Clarke, in his “Irish Copyright and Design Law” considers the provision only insofar as it applies to a degree of reproduction, stating “It is clearly an infringement of copyright to put a work on an internet server without the consent of the owner of the work”. This would include activity like uploading a film to Youtube. In the case of linking, no work is placed on a server; rather, the reader is directed to the copyright owner’s own server. In Clarke & Smith’s Intellectual Property Law in Ireland, Section 40(1)(a) is explicitly described as specifying “that placing works onto a server – acts of uploading – will be infringing acts”. The furthest the authors are prepared to go with regards to linking is to state that
“It may be possible to view the “authorising infringement” concept to be so wide as to discover acts of facilitation such as providing information on the location of infringing works on the world wide web, even if the infringing work is not available directly from that website”
The emphasis here is not on the act of linking, but rather the target of the link. It would follow that where the target to the link is not infringing, the link itself is not an infringement. The “making available” of the material is done by the newspapers themselves. Any external link is simply a direction to potential readers that here they may enjoy the article, much as one would direct a potential audience to the performance of a theatrical work by notifying them of the time and place of the performance. Contrary to their earlier statements to the contrary, NLI now appear to agree.
In recent days, there was much emphasis placed on the Terms & Conditions of individual Newspapers. Prior to NLI’s change of position, the Irish Times (whose responsiveness in this matter puts their fellow newspapers to shame) pointed to the provision in their T & C’s which permitted linking, for personal, non-commercial purposes. This of course assumed that such permission was the Irish Times’ to grant. The reason for this distinction was, in all probability, that the newspapers were attempting to reserve their position with a view to a possible license claim against google. The new position of the NLI seems to render any permissions granted in the T & C’s moot. Time will tell what has become of the newspaper industry’s aspirations regarding Google and other search engines and aggregators.
If, as is now clear, you can link without permission, what can’t you do? Section 39 of the 2000 Act provides for the “Reproduction Right”:
“There shall be a right of the owner of copyright to copy a work or to authorise others to do so which shall be known and in this Part referred to as the ‘‘reproduction right’’”
Copying or transcribing an article is clearly a breach of this right. How much of an article one can get away with quoting is unclear, but there is Court of Justice of the European Union case law to the effect that the creation of an eleven-word extract from a news article was capable of constituting “a reproduction, in whole or in part”. Thus, a short extract, where “the elements thus reproduced are the expression of the intellectual creation of their author” must come under the protection of the Reproduction Right. Technically, then, even reproduction of a headline might require a license.
An interesting side note, and one that did not receive much attention in the recent debate, is the question of who owns the copyright in the work of freelancers. Section 23 of The Copyright and Related Rights Act, 2000 provides
“The author of a work shall be the first owner of the copyright unless—
(a) the work is made by an employee in the course of employment, in which case the employer is the first owner of any copyright in the work, subject to any agreement to the contrary,
The effect of this provision is that newspapers are the first owners of copyright in the materials they publish, but only where the author of the article is an employee. Where, as is very often the case, an article is written by a freelance journalist, the journalist retains copyright unless he specifically waives it. Given the manner in which freelance work is commissioned, there is rarely any explicit agreement between freelancers and newspapers as to what rights are being acquired by the newspapers.
In the event that Newspapers have made demands for payment in respect of material in which they do not hold copyright, there may be civil liability for an account of profits, as in the US case New York Times Co. v. Tasini, where the Plaintiffs, a large group of freelancers, won a compensation pool of $18 million from the Times. The spoils of any such case in Ireland are unlikely to be so rich, but it’s an issue one would expect the NUJ to look at on behalf of its freelance members.
Update: A site which linked to this post did so in the context of a discussion regarding the meaning and limits of “Fair Use” in Irish law. By way of clarification, Fair Use has no meaning in Irish copyright law. The idea that it does is probably (along with the "Class Action" canard) the most commonly repeated item of legal misinformation online. Fair Use is an American doctrine with no equivalent here.
Fergal Crehan BL 
Hi Fergal,
Excellent post and very good. I agree with the distinction on “linking” and “deep linking.” Deep linking is just a term used by web designers and, largely, Google and often, special interest groups like the SEO community, to distinguish and expand our own vernacular.
My issue with #linkgate – and why I’m dissappointed that so many have accepted the Irish Times’ half hearted disingenuous backtrack – is that its about indexing that data for use by search engines and/or news portals.
What’s funny is that the Irish times have 8,500 outbound links to different domains. If each link is to a unique domain, they are, by their own daft idea, liable to up to some €2.5 million in fees.
Whatever the argument may be made about “Fair use” – to use a small part of copyright content (feel free to clear that up) – The Irish Times (and the other NNI/ EU Publications) claim that they don’t want to be indexed (or scraped, as they refer to – similar to Ryanair -v- SkyScanner) – is that they put so much work into making sure that Google does index them and even rank them.
Hi David. You rightly say that SEO is something news organisations work on. However, a problem can arise when third parties move beyond simply referring users to news organisations’ sites. The newspapers want to protect themselves from third parties using industrialised reproduction of newspapers’ material in order to keep people on their own services, and away from the newspapers own sites. From a newspapers’ perspective it’s understandable to court inbound links on the one hand, while acting against industrialised reproduction on the other.
But you’re not “acting against industrialised reproduction.” In fact, the Irish Times is doing the exact opposite, it’s requesting it.
I’ve randomly selected an article from your site: http://www.irishtimes.com/newspaper/breaking/2013/0108/breaking12.html
When I request the page source I find this robots meta tag:
For more detailed information, you can see this documentation: https://developers.google.com/webmasters/control-crawl-index/docs/robots_meta_tag . It exists elsewhere, but to be honest robots.txt is one of the web’s more de facto standards and the link I gave sums up the various docs quite nicely (specifically the table in “Valid indexing & serving directives”). However to summarize quite simply you can do the following on your page instead:
Now automated web crawlers will not have cached copies of your pages or show snippets. You will still show up in search sites. Your links will still show up on well-behaved news aggregation sites. None of your content will show up. If it does, they have not only violated your English language copyright, but your robots meta-tag directive.
As it stands, you tell web crawling robots to do anything at all with your content. You explicitly do that even though that is the default on the web.
If you’re going to publish on the web, could you *please* learn the technologies involved before trying to lobby for legislation. Believe it or not people have actually considered your rights and they have actually implemented systems to protect those rights. You are choosing not to use them. In fact, you are using those systems to say the exact opposite of what you want.
Amusing. The HTML markup I was trying to show was removed. Here it is without the angle brackets:
This is the robots directive that the Irish Times appears to put in all their articles (I didn’t check them all): meta name=”robots” content=”all”
Based on their policy, this is what they should have: meta name=”robots” content=”nosnippet,noarchive”
Not a solution for the issue at hand.
Good post, further to the fact that every link generated within a domain remains on the table of the server and is in the gift of the server(s) (either physical or virtual depending on hosting arrangement) to be withdrawn, to redirect, or manipulated or simply not be served, the domain itself is granted to the content owners by the domain name registry authority. A domain on a TLD is not held in perpetuity or in accordance with copyright term length. It is possible to stop someone registering a domain that infringes a IP or uses a real person’s name etc, but one isn’t entitled to a functioning domain because one has a copyright claim on the name or entitled to it for a period normally allowed for protection of copyright under relevant laws. DNS servers don’t have to organise copyright licensing as far as I’m aware of, do they?
Also as David points out there are plenty of outbound links from the Irish Times, a commercial site, referring to resources deep within other websites. SSome portions of the Irish Times website (and indeed the other papers) rely on curation of links that is hard to distinguish from aggregating websites. For instance blogs dealing with music are linking to 3rd party content directly and I doubt very much the authors were told to clear copyright to such links. Because that would be daft.
Hi Ronan,
I made the very same point last Friday when asked about irishtimes.com and linking. Our site has thousands of links to other sites, including the blog ones you mention. And we see no need to be licensed to do this. So to require other sites who link to us to acquire a licence would be… well, on Friday I said hypocritical, but on reflection I agree daft is a better word.
it was not just 11 words that were the problem but 11 words extracts on indsutrial scale
I’m curious how “community standards” might come into play. In particular, how does it relate to reproducing or extracting snippets from content?
Things published on the web need to conform to a number of standards in order for the content to actually be successfully published. There’s the HTTP standards for the transmission of the articles, the various HTML, CSS, Javascript, image format and other standards needed to be conformed to such that the article can be rendered.
In addition to all of that, web technologies have always had some form of discovery standards. In the end the primary one relates to managing the automated crawling of web servers which is the robots.txt standard. Included in that standard is a way to allow or prevent automated indexing, snippetting and caching content.
As an example, the Irish Times robots.txt file restricts automated crawling of portions of their site but does not restrict access to articles. In addition, the Irish Times contains a robots meta tag that explicitly allows web crawlers to cache, index and make snippets of their articles. This explicit tag is actually not even required as that’s the default on the web.
Essentially, the published copyright terms (in English) do not conform to the published copyright terms (in code). Which takes precedence?
Pingback: Irish Newspapers Mine Links For Revenue | ImpressiveNews
Pingback: Irish Newspapers Mine Links For Revenue « Gadgetizing
Pingback: Irish Newspapers Mine Links For Revenue « MattsLens
Pingback: - Media Focus uk - Strategic Intelligence in a Digital World
Pingback: Do Iona Own a Tweet? | Fergal Crehan BL
Pingback: After The Hyperlink Hype | Broadsheet.ie